Enterprise Processes Behavioral Analysis Using IDEF3x and Bayesian Network

Analysis of company processes behavior for its performance evaluation is the natural response to take into account both the expected states of performance and those of whose uncertainty is partial if not total. For this analysis, we use the Bayesian networks which we build from a reference model of the process represented by IDEF3x. The method to translate IDEF3x model to the Bayesian networks is introduced in this article. Here we discuss the choice of IDEF3x, we present the stages which we follow for the development of a probabilistic model of the studied process, we explain how we identify the various flows of the process on which the graphic passage of IDEF3x to the Bayesian networks rests. To illustrate our approach, we took an academic case integrating various possible configurations of a process. The approach developed can be applied to service systems